It’s Halloween, y’all! ‘Tis the season for scary movies, fun costumes, and of course David S. Pumpkins.

We have some frighteningly spooktacular new features we’re just dying to share with you. Seriously, they’re so amazing they’ll haunt you. But in, like, a Casper the Friendly Ghost way, not an Amityville Horror way.

This feature is all treat and no trick. About 10% of PDFs uploaded to Logikcull pose a particularly thorny challenge: Primarily image quality degradation and, consequently, loss of existing searchable (and indexable) text. So for instance, if you uploaded a PDF that included an embedded screenshot of a text message, Logikcull’s OCR might pass over that text for lack of image clarity. The question is, what previously un-indexable text is lurking in those PDFs -- and, on a severity scale, are we talking, like, Mike Myers… or Michael Myers?

Well, at Logikcull, we strive to give you the fullest picture of the evidence. So just as Laurie Strode refuses to stop fighting The Shape, our engineering team did not back down against this OCR challenge. They fashioned a weapon out of a wire hanger and fought back. And thus Deep Text Recognition (DTR) was born. (I’m 99.9% sure this is how software engineering works.)

DTR extracts deep text from embedded images in PDFs, preserves existing searchable text, and prevents severe image quality degradation. It also expands OCR to 100% of PDFs uploaded into Logikcull. This means that when you upload a PDF into Logikcull, even if it only has images of text, that text will be found, indexed, and made searchable.

When you upload, Logikcull will automatically run DTR on PDFs. Then it will apply a “Has Deep Text” QC tag to any files it finds with deep text. This is your heads up that the PDF has an embedded image containing text.

You’ll also be able to see a new field in upload stats called "Deep Text." This represents the total number of documents with that deep text QC tag.

No more dark data hiding out in your PDFs means one less thing to be scared of.

There was a time when Logikcull users who wanted to upload multiple files had to manually zip them on their computer before doing so. This can be a confusing pain if you don’t know how to zip files, which a lot of people don’t. You may or may not be reading a blog by one of those people right now.

When you’re not familiar with something, it takes extra work to deal with, which means valuable time could be wasted. Think of The Thing: All that time they spent researching the mysterious extraterrestrial that was imitating and killing their coworkers was time they could have spent doing their actual jobs researching Antarctica (or whatever).

Attorneys and paralegals have better things to do with their time than figuring out how to zip files. There’s also the risk of creating numerous, burdensome individual file uploads because of the confusion around how to zip.

Well, good news! After some time conducting alien autopsies and poking blood with hot wires, our engineers have made Logikcull zip your files for you.

Now all you have to do is drag and drop all those files into Logikcull, and Logikcull will zip them up automatically.

And pay-as-you-go users will only be charged on the compressed, uploaded size of the data. More pocket change for that post-Halloween sale candy.

When sending a legal hold notice to someone, it can be really helpful to include a survey to learn more about the custodian and their data. Doing your due diligence and getting to know someone a little more before you deal with them is important. Maybe if the kids in Carrie had bothered to get to know her, they wouldn’t have (spoiler alert) dumped pig blood on her at prom and gotten themselves into that whole unfortunate situation at the gym.

We can’t guarantee that using Logikcull’s new legal hold surveys will protect you from angry teens with telekinetic powers, but they can help you manage the scope of your holds.

Creating a survey is as simple as checking the “Include a Survey” checkbox when you create the hold. Then hit “next” and you’ll be taken to the page where you can customize your questions.

From there, you can add questions and answer types—from multiple choice to paragraph responses— and rearrange questions however you please by dragging them.

The recipient doesn’t need a Logikcull account to complete the survey.

For all the details, including a demo video, check out this article.

I have to admit, as I start writing this section I don’t have a good horror movie equivalent for it— though calling the previous ones “good” may be a stretch anyway. We’ll see if I come up with one to shoehorn in as we go along. That’ll be fun for all of us.

For those who may be unfamiliar with Markdown, it’s a text-based file format similar to HTML that’s used extensively in products like Gitlab. Just a few short weeks ago, Logikcull didn't support Markdown at all, so if you were trying to use us for Markdown, you were as out of luck as a teen at Camp Crystal Lake on Friday the 13th (HEYO!).

Now, Logikcull supports text extraction to provide the normal search experience. Logikcull also provides a lovely rendered version of a Markdown file similar to what a user would see when using a product like Gitlab.

And just as a note: Judgment calls were made on how to present certain aspects of these files in PDFs. If any users have thoughts or comments on how Markdown is presented in PDFs with Logikcull, please let us know! Customer feedback is the best way for us to keep improving, and we always love hearing from our users.

HAPPY CULLOWEEN!